Friday, June 12, 2015

What Is The Value of a Break?

So what is the value of a break?  Every point in ultimate is worth one point in the game, so it seems like the answer to that question you be "one."  But in the elite game we all expect offensive holds and cherish breaks.

I wanted to get a better sense of breaks because I was looking at Draymond Green's plus minus rating from game four of the NBA finals and adjacent to the stat was his real plus minus.  While we can easily look at an ultimate players normal plus minus (stat heavy leagues like the AUDL have those values on their stats pages already), but how can we adjust those numbers?

I guess we could do some sort of heavy adjustment talking about when the break happens, but that seems a step too far right now.  Let's just set a simple bar and see if people tear it apart in the comments.

First, it is silly to think about most elite ultimate as single point endeavors.  I don't have the numbers in front of me (this is part of a different project for this summer) but if the Possessions Per Goal (PPG) metric for a team is greater than two it means they will need two possessions to score and that somewhat means that a single turnover is likely on any given point.  I ran the college numbers for ~12 games this season, and it seems like the most elite college teams can operate at sub 2.0 PPG, so let's assume that club teams can do similar or better. What that leaves us with is that a turnover in a possession is still unlikely (again, data unlikely) which holds with our expectation of club where offensive holds are expected.

With many elite club teams operating with an offensive line and a defensive line it makes more sense to think of a "round" of ultimate as being a two point exchange.  At any given time the actual "score" of the game should reflect that exchange, and one way to do that is to include half-point into the scoring system.  For example, if you receive the pull and throw a goal making the score 1-0, the score is actually +0.5 (where the plus indicates your team's value, it would be -0.5 for the opponent).  Upon the opponent scoring it would be driven back to +0.

Perhaps this comes from a mental game as a coach, where you tell your team that a lead isn't as big as it thinks it is.  Being up 4-3 but pulling means you aren't up at all.  But at the same time the half-score offers something that just pure breaks doesn't allow, an indication of who will win.  True, games are won by the number of breaks you get.  Sure, if you get more breaks than your opponent then you will win the game.  But if we just track breaks then who wins with a score of 0?  The team that wins the flip!  So there is good cause to think that is terms of game states the flip is worth +0.5.

Game state isn't the same as plus minus.  The point of looking at the value of the flip is that it shows you the value of the change of "serve" of the game: +0.5.  So if we treat an offensive hold as +1 (which seems like a good baseline) then we can argue that a break should be worth +1.5.

That is a simple formula.  A players plus minus is: (O-points scored+1.5*D-points scored)/points played.

I suppose we could go with a more mathematically complicated model where we scale the break multiplier based on the number of breaks in a game.  Something along the lines of

(O-points scores+(1.5-0.1*Game D-Breaks Scored)*D-points scored)/points played

This would reduce the value of your break based on the number of breaks in a game.  That kind of makes sense, and perhaps allows the statistic to be used on a wider level of play.  In games where offensive holds are difficult the value of a break would eventually get below 1, making it less valuable than an offensive hold.

So what are we trying to do with this statistic?  The number can only fluctuate between 0 and 1.5 and would seem to measure the value of a point with a person on the field.  Odds are your best offensive players are going to be just above 1.0 while a defender with a number greater than 1.0 is a godsend.

I've got some film to watch to look at PPG in the club division.  Maybe I'll go through some college games and track this or track it while I am following a few club games.

Thursday, April 09, 2015

Looking at numbers

I have a job for the Atlanta based AUDL team the Atlanta Hustle.  I think my title is technically "advance scout" which basically means I'm going to look at film and try to give assessment docs to the coach so he can map out practices and strategy.

Anyway, the GM is a numbers guy, so I think I'll need some numbers/video to back up my assessments.  Good thing the later is my strong suit.  The former I need some work on, so that is what I have been doing recently.

I have a poor relationship with ultimate statistics.  I loved the work that Ultiworld did a long time ago through tracking every flipping pass in NexGen games.  But that app is dead, and I don't really know to do with a chart that shows me a team's scoring probability based on field position.  I guess it could expose that a team is particularly weak in the coffin corner, but in general I don't know what else I can get from that information.

So this high school season, since I am a sub-called, I decided to play around with statistics for the team.  We haven't been having our most successful year so I started tracking the number of unforced errors and the number of possessions.  It hasn't really righted the ship, but it is interesting what the numbers show.

First a bit on "unforced errors."  In order to get around the tricky subjectivity problem I decided that if the defense doesn't touch the disc it is an unforced error.  There are some strange things that fall into that category that make the numbers hard to really analyze.  For example a punt in the wind is considered an unforced error, as is a high stall punt that no one touches.  Also a jump-ball that the defense doesn't actually touch (but they clearly influenced the play) counts as an unforced error.  It would make sense to fix some of these issues, but then we get into the subjectivity of "was that a punt or a huck too far?"  Or "how much did the defender actually influence that drop?"  These are things that I want to avoid so I'm keeping it pure.

Basically unforced errors are possessions that end with you giving away the disc, which means you aren't making the defense take it from you, which is a bad thing . . . right?  So I took the number of unforced errors,  divided by the number of possessions and we now have UE% which tells us the percentage of possessions that end in unforced error.

For Paideia that number was frighteningly high (~50% or more at times) which pointed out how we were really beating ourselves.  If we could improve on that number then we would at least be asking more of our opponent.  I don't want to get into Paideia's season, but it has been having a good impact. One other thing I was able to track was what I am calling our "Conversion Rate" which is just the number of possessions divided by the number of scores.  An average of this over multiple games tells us how many times we need the disc (on average) in order to score.

Here is where I feel like I got into something that was useful, tracking possessions.  In the past many teams have been concerned with offensive holds and defensive breaks.  But a defensive break that requires 4 possessions to score isn't the same as break that only takes one.  I think moving away from line-based statistics and moving toward possession based statistics will offer some new insights to analyzing the sport.

After playing with this for a little bit I wondered what was a reasonable UE%?  Did it change per level of play?  So I set off looking at college games from this season.  I have made it through just over 20 games and here is what I have found.

The average UE% of the games that Ultiworld has filmed is around 30.79%.  The funny thing is that the average for a winning team isn't any lower than that of a losing team (30.75% vs 30.83%).  What is even more interesting is winning the UE% battle isn't a good indicator of success.  Plenty of teams have won their games despite having a worse UE% than their opponent.  I guess this would speak to the idea of there being "good" turnovers.  This metric still gives us a glimpse of how many times on average a good/elite college team will just give you the disc back.  Looking at the similar numbers for the college women's game and the club games might provide more support for what we assume (better levels of play make fewer "mistakes").

The real insight came from conversion rate.  First of all, looking at the conversion rate of a single game is boring.  Because possessions for each team are never more than one away from each other if you win the game you won the conversion rate.  This is one of the things I hate about certain statistics like "breaks" and "turns."  Guess what, if you get more breaks than your opponent you won the game.  If you commit fewer turns, you won the game.

But this metric did offer some insight over a number of games.  For example, there seems to be a clear line between the best teams and the next step down.  Elite teams (Pitt, Oregon, UNCW) have a conversion rate that is typically sub 2.00.  Other teams tend to operate above that mark, with some of the worst being as high as 3+ (which is where my high school team operates at times).  It is no surprise that the average for winning teams would be lower than that for losing teams.  Conversion Rate basically tells you the number of possessions you need to score (on average) so winning means a lower number. The winning teams have an average of 2.06 while the losing teams have an average of 2.47.  There are some teams that lose with a CR below 2.  Those are typically good, or at least efficient, games.

What I'm curious about now is how good a predictor average CR for a pair of teams is for the game's outcome.  In general does the team with the lowest CR win future games.  How should standard deviation of CR play into that calculation.  Washington has a poor CR (2.41) but was able to post one sub-2.0 number.  Could they get hot and beat an elite team by putting up their best efficiency number (1.83).  Pitt (1.76!) has a fairly stable CR, so the likelihood of getting a "bad" game out of them seems low.

I feel like there is some room for innovation there.  Given enough data we could look at the effect a "good defensive team's"impact on opponent's CR as compared to that opponent's average CR.  Anyway, I have to go.  Hopefully I didn't ramble too much.

The re-emergence of the pivot reset (Incomplete)

Two years ago I would have argued that I watch about as much film on ultimate as anyone with a job (there was no way that I could keep up with some high school kid's youTube binging).  That isn't really the case anymore with the proliferation of film through the now defunct NexGen, Sky'd and Ultiworld.  I still watch a lot of film, but I find less of a need to spout off about things that I see now that there are other (arguably more capable) people doing that.

But watching the Stanford women's team's games from this year's Stanford Invite I saw the re-emergence of a move that has been building for a few years, and so I feel the need to point it out.

When explaining a vertical stack I will at times refer to the person at the front of that stack as the "Pivot."  The idea is that when the disc swings from one third to the other the top of the stack is the rotation point (pivot) of that swing.  We all know what a reset is.

The pivot reset is by no means a new idea.  I first learned of it over a decade ago from a former Godiva player, who used it over a decade before that as the primary reset style.  So let me have a brief history lesson while understanding that despite playing for 20+ years I'm talking about things older than myself and am bound to get something wrong.

There was a time when the concept of having a player behind the disc seemed ridiculous (at least to me).  We want to move the disc forward, so that is where the people should be.  If a person couldn't get open by 10 (which was longer back then) then we would punt.  While my team was stuck punting some teams had figured out how to reset the count from the front of the stack.  One of those teams was Lady Godiva.

The idea of Godiva's reset structure was that the first two people in the stack were the resets.  If flow couldn't continue those people would run a specific pattern.  This involved cutting forward to break your defenders positioning (either by turning their hips or getting them to backpedal) then a 90 turn to either side.  The next person would then do the same to the other side.  Weaklings would default to the open side, which would still work often.  But Godiva players were good at throwing the around so they would use that break often.

This isn't exactly a pivot reset, but lays the groundwork for what happened almost 20 years later.  The vertical stack changed drastically in the 90's (Tiina, help me out here and tell me how wrong I am).  We moved resets behind the disc because there was more space there and the forward reset died off a bit.  Then horizontal offenses became en-vogue and the whole structure of resets changed for a while.
Some college teams, and maybe even some club teams would run a front of the stack reset, but it had moved out of the zeitgeist.  As few as 6 years ago I started noticing more New England college teams running it, perhaps as Godiva players started coaching or because defenses became bad at it.  There were evolutions to the system, and it didn't much look like what Godiva ran decades before.  (This was probably a good choice because what Godiva ran required solid discipline and excellent timing, which are not abundant in the college division).

Instead teams were often running it with a backfield reset as well.  The backfield would serve as a primary, and if that wasn't working then they would shift to the front. This made tons of sense and was a logical continuation of a horizontal reset scheme where you look at one side and if that doesn't work you check with the other side that is running counterflow.

Those college players eventually move to club, and club defenses get used to particular offenses, so lo-and-behold we can see club teams using this strategy more and more.  But still it was more regulated to the northeast than other schemes.  Ironside has been using it more and more over the past few years.  It is difficult to defend against, especially with the increase in off-hand throws and creative breaks.  The pivot defender has to choose between overplaying the breakside to prevent this reset and allowing the openside pass, or play to the openside and risk being exposed by a (tight-windowed) break.  I'll get into defending this later.

But the thing that convinced me that this is coming back was Stanford's use of this system.  This is a west coast team utilizing a scheme that is entrenched in northeast history.  So apparently someone was paying attention and soon other teams will be as Stanford's women's team looks pretty darn good right now.  Despite being understaffed from a roster standpoint, Stanford hung with powerhouse Oregon by using this reset scheme.

In the clip below (sorry for its length) Annee Rempel (#15) is playing the pivot through a series of throwers.  When she gets the disc and throws it, she clears back to the front.  When the thrower is in trouble she knows to expect a throw to the break side.  She even bites on a Stephanie Lim breakmark flick, thinking that it might be for her.




Is this a model for teams that are under-athletic to compete against large schools and their hulking athletes?  Not so fast.

First let's analyze the modern iteration of this scheme.  The primary reset is still behind, but you have a secondary reset at the front of the stack.  When the disc is on the wide side of the field the pivot has the skinny lane as an option (something similar to a pocket pass in basketball).  When the thrower looks at the pivot they then get the option of juking to get open.  When the disc is trapped this system doesn't work as effectively, but you still can use the pivot to replenish a backfield reset when the original reset cuts downline.

One thing that is particularly difficult about this reset is the possibility of the pivot post.  A "post" is basically the same as a post-up entry pass.  The pivot is in position and the thrower is just going to put it to a space knowing that the pivot has best line of sight and should be able to get to it first.  This post happens often on a backfield reset when the defender is playing straight up.  But a pivot post is incredibly difficult to defend without losing position on the pivot.  In order to prevent the post you need to be able to see the disc, but you can't see the disc without at least turning 90 degrees on the pivot, which gives them an advantage towards one direction.  Due to the short length of the throw, that isn't much time for a defender to recover from that advantage.  So to take away the advantage the pivot defender will play straight-up (faceguarding), which opens up the post pass.  You can see the issue.

This is why it was working so well for Stanford.  Looking at their percentage of unforced errors per possession in the finals against Oregon it was a jaw dropping 29%.  Much of that was because Oregon was getting blocks, but it also means that Stanford was only giving away the disc only three out of ten times.  By contrast Oregon's %UE was upwards of 43% (mostly on long hucks).  Stanford was able to keep possession and in large part it was due to this reset system and excellent handing.

And that is really the key to this scheme, you need to have great handlers.  They need to have some good throws to tight spaces (such as Monisha White's use of her lefty backhand) and they need to be able to lead a pass well when needed.  Your resets (both front and back) need to be able to stay close without clogging lanes.  If your back reset creeps in too much they allow their defender to sag into the throwing lane without being punished.  If your pivot reset shifts to the openside too early they alert their defender while also crimping the openside pass.

So I don't think this is a scheme that just any team can work with and expect success.  But if you are a team with some incredibly strong handlers with good vision and chemistry (looking at you Texas A&M) this becomes a viable structure that can allow you to hang with more athletic teams while saving your legs for sunday.

Wednesday, October 22, 2014

2014 Club Nationals "Mobility"

The exploits of Ghetto Birds/Rhino coupled with the travails of Machine/Polar Bears made me think about the delta between a team's initial seed at nationals versus their final seed.  So here is a quick compilation of that data for 2014 Nationals.  In the case of ties (3rd place, etc.) I'm following end of season USAU Rankings as the tie-breaker.

Open:
1.   Johnny Bravo (+4)
2.   Ironside (+2)
3.   Ring of Fire (+3)
4.   GOAT (+6)
5.   Revolver (-4)
6.   Rhino (+1)
7.   Chain Lightning (+2)
8.   Truck Stop (+6)
9.   Sockeye (-6)
10. Doublewide (-2)
11. PoNY (-1)
12. Temper (+3)
13. Furious George (+3)
14. Machine (-12)
15. Prairie Fire (-2)
16. Sub Zero (-5)

Womens:
1.   Scandal (+3)
2.   Fury (+2)
3.   Brute Squad (-2)
4.   Riot (-2)
5.   Ozone (+3)
6.   Showdown (+4)
7.   Traffic (-1)
8.   Nightlock (-1)
9.   Molly Brown (-4)
10. Schwa (+2)
11. Heist (+0)
12. Underground (+3)
13.  Nemesis (-4)
14. Capitals (-1)
15. Green Means Go (-1)
16. Tabby Rosa (+0)

Mixed:
1.   Drag'n Thrust (+1)
2.   Seattle Mixed/Ghetto Birds (+10)
3.   The Chad Larson Experience (+0)
4.   Wild Card (+5)
5.   Mischief (+0)
6.   Blackbird (-2)
7.   Bucket (+1)
8.   Santa Maria (+6)
9.   Slow White (-1)
10. AMP (+0)
11. Polar Bears (-10)
12. 7 Figures (-6)
13. American Barbecue (+0)
14. The Administrators (+2)
15. Cosa Nostra (+0)
16. D'Oh! Abides (-5)

If you take the average of the absolute value of the delta then you get a measure of how much the seeding held up.  We aren't going to take a sum of squares o that larger movements aren't more heavily weighted, but we need to take the absolute value because it is a zero-sum game.  Let's call that metric "Mobility."  Here are the mobility numbers for the 2014 Club Championships:

Men's: 3.6875
Women's: 1.9375
Mixed: 3.0625

Since all of these numbers are averages it is kind of interesting that Men's had the most Mobility.  Women's being the least isn't surprising, and Mixed having a "high" number is totally not surprising given the rampant turnover in that division.

Machine's drop into Elite status seems fuel the Men's division Mobility, but it isn't that atypical for nationals.  While -12 is the largest number in the past five years (in the Open division), there have been other numbers as high as +8 and even -10.  If we look at the numbers in the Open division for the past five years we see that the new format doesn't automatically produce more Mobility, and that this year might have been extreme but isn't that far past previous years values.

2013:  1.9333
2012:  3.1875
2011:  2.8750
2010:  1.7500

Ignoring last year (when things were pretty much chalk and we had a new format) there as been a five year increase in Mobility in the Open/Men's division.  Now, it could be the result of more parity as the talent base has increased over the past five years.  Anecdotally, there are fewer teams from the same city attending the Club Championships than there were in the past.  That speaks both to a consolidation of talent in those cities and to an increase in talent in other cities (necessitating that consolidation).  Let's see what has been happening in the other two divisions:

Women's:
2013:  1.7500
2012:  2.7500
2011:  2.2500
2010:  1.6250

Mixed:
2013:  2.6250
2012:  3.3750
2011:  2.2500
2010:  2.6250

These numbers look like the hold up better to our expectation.  The women's division is on average less Mobile and while the Open/Men's division can have its crazy years, it is less Mobile than the Mixed division.  What does this all mean?  I have no idea.  Mobility hinges on the ability to correctly seed the tournament as compared to the final outcome.  I don't know if that is actually the goal of USAU.  Poor connectivity plagues the Mixed division, which would impact Mobility.  But even in the more highly connect Open/Men's there is still a reasonably high Mobility.  I guess that speaks to parity?

It could also relate to who takes placement games seriously.  The difference between 12 and 13 is just as significant as the difference between 2 and 3 under this metric, but I imagine those games are not necessarily played with the same level of intensity.  Out of curiosity I looked at the Mobility for the top and bottom 8 each year in the Open/Men's division.  Here is what I got:

               Top 8        Bottom 8
2014:       3.5               4.25
2013:       2.125           1.625
2012:       2.875           3.5
2011:       2.375           3.375
2010:       2.875           2.875

Aside from 2013 (which we already said was a chalk year) this shows what we would expect.  The lower seeds tend to have more Mobility than the higher seeds.  Could this be because teams let their foot off the gas in placement games?  Could it be because there is less connectivity amongst the lower seeds so correct seeding is more difficult?  Or could it be an increase in parity in the second tier (if not in the first tier)?  I have no idea.  I was just bored when I though about how far/high some teams fell/rose this year and decided to figure some things out.

Thursday, August 07, 2014

The power of the かいひステップ

I just woke up this morning to literally see the final score of the Buzz Bullets vs Ironside prequarters game at Worlds.  Buzz shocked everyone (or at least me) by winning 17-16.  Since I wont be able to actually watch the game for a bit I can't really give a good analysis, but while watching the highlight reel I noticed something that Scion Scone wrote about in an ultiworld article not too long ago.  He commented about the way the Buzz generates hucks to odd spaces.  We've known that was the case for a while, but I've never really looked at how or why.  Usually my time watching Buzz is about seeing how well they get short breaks and their strange, no-mark zone.

Then watching the highlight reel from the Worlds game I noticed something.  Prior to throwing those hucks Buzz turns the wrong way.  To further explain, when a player is cutting and catches the disc the direction they turn is typically the shortest path to facing upfield.  On an slanted openside cut that means turning over the openside shoulder, on a breakside cut that means turning over the breakside shoulder.  This has a great natural feel and can make for quick disc movement because the player's momentum expedites the turn rather than hindering it.  I can go on for a while on this, but the important thing is that most cutters turn this way.  On all of the dangerous hucks (the one Scion is referencing) Buzz turned the wrong way.

So why does this matter.  Let's imagine a pass going to the open side and a defender trying to get the block (as many Ironside defenders were trying).  That defender, especially if it is a well versed Ironside defender, is going to start on the open side and try to fight for open side positioning.  That way they are closer to the block without just being flat out faster than their defender.  But there is a disadvantage to that tactic.  If you don't get the block you are out of position to set the mark because you are on the open side.  This is a common problem that happens all the time in the States, and teams try to exploit it.  But the amount that you can exploit it is limited if you are still turning with your momentum

Again, think about this on an openside cut.  If I catch the under and then turn to the open side, my ability to throw a huck to the opposite third is limited.  Let's say it is a backhand force, and I catch the disc on the strong side.  I turn to throw a backhand and realize that my defender has over-commit to that side.  So I've got a wide open flick huck, but my turning momentum (not an actual physical term) is the wrong direction.  In order to get power on that flick huck I will have to pivot over and general new momentum the opposite direction.  That takes time.  While I wont be marked, I will likely feel pressure and my receiver is getting farther away by the minute.

But what if I turned the other way?  Then my momentum is turning the correct way for that huck.  I can drive it farther, and do so faster than if I turned the other way.  The speed on that delivery is really key because it can further exploit someone in the same third running away from their defender.  If the deep target's defender is on the open side (which they would be, right) then they are out of position for this breakside huck.  If I can get the disc out quickly they wont have time to recover.  If I pivot the wrong (or actually the right) way then it takes longer and the defender has time to recover (or get sideline help).  Think of it as a turn that gets away from your defender.

So the かいひステップ (Kaihi-step: Kaihi means Avoidance . . . I think), is turning the wrong way on an under cut to push the active space back in the direction (laterally) from where the disc came.  Should it be used all the time, no!  But using it in a spread scheme when the disc gets near the sideline seems like a potentially good idea.

I'll go through and look at the film soon and try to follow up with some video, but I wanted to get this idea down on "paper" before I forgot it.

Wednesday, July 23, 2014

Practice efficiency: drills vs scrimmaging

I felt bad that Martin was the only one posting here these days, and since I've recently gotten back into coaching I figured I should try to get back to posting some...

While setting up a plan for Southern Revival's 2014 season there were a few restrictions I had to keep in mind. We were incorporating a bunch of new players and we didn't have a ton of practices (3 practices before the series) so the efficiency of practice became really important. Traditionally I tried to break the game down into the simplest level possible then gradually build up complexity. So maybe we'd start with footwork while pivoting and then add in a disc (without throwing), then throwing, throwing with a mark, throwing to a moving cutter and eventually scrimmaging. The idea being you focus on the individual pieces then put them together.

However, I'd read recently that this 'common sense' approach to learning was flawed. Instead of focusing on individual skills, you should work on multiple things at once. That way your overall skill increases more, even if an individual skill may not reach the level it would if you focused solely on it. Also, working on multiple skills together you improve the coordination of the skills together.

So, I ditched drills (almost) completely in favor of scrimmaging. Practice consists of a brief warmup which may include a drill or two, then scrimmaging broken up with throwing drills (no running) and/or whiteboarding time to allow for recovery between games. The games themselves have different rules to try and focus on different skill sets we want to work on. I'll stop the scrimmage occasionally to point out what we should do in particular situations, and frequently pull individual players aside for some one on one coaching. Players also keep track of how many games they win throughout the day and when we run sprints at the end of practice they can subtract their games won from the total sprints.

Here are some of the games I pull from when planning practice:
  • 3 on 3
  • 5 on 5
  • 7 on 7 with shorter stall counts
  • 7 on 8 (man or zone d)
  • 10 (or 5) pull
  • extra points for specific tactics (breaking the mark, dump/swing, etc.)
  • double score
  • redzone scrimmage
  • start with the disc trapped, start from deep in own goal, etc.
It's worked out fairly well so far. My biggest concern was having enough bodies to scrimmage but we've had fair attendance so far so it hasn't really been an issue. Number of touches was also a concern. Some players are naturally going to be less involved in full 7 on 7 games whether because they are less skilled or new to the team, and I was worried that this might slow development for those players. To help combat this, we run a fair amount of 3 on 3, particularly early in practice to get everyone involved. I also keep an eye on involvement as practice goes along. 

I do think drills have a place and we'll occasionally have voluntary skills/drills practices where we typically don't have enough to scrimmage and we focus more on individual skill development. However, I think it's still important to try and make those drills as game-like as possible and try to incorporate multiple skills into each drill.

Monday, July 21, 2014

Three Ways An Ultimate Team Gets Better In The Off-Season

How's that for a click-bait title?  I'll go back and change it, but I think is still pretty fitting for the topic at hand.  This is something that had been mulling in my head since sitting on Chain's meeting about the upcoming season last Winter.  I was listening to a lot of basketball/football podcasts over the summer and I think I may have heard Bill Barnwell mention these three things at one point or another.  Maybe not, though. Fortunately, I got a wall to bounce some ideas off of (thanks Jeremy Goecks) this past weekend at Master's Nationals and I feel a little bit more confident in the ideas.

Basically there are three ways to improve your ultimate team during the off-season.  Those would be drafting, development and free-agency.  To add more detail:

Drafting:  Getting college players previously on a weaker team or not playing in the club series to join your team.  This could also be the case for out of college players, but the idea is that they aren't really part of the elite-club ecosystem yet.

Development: Improving the skills and athleticism of the players currently on your roster.  Anything that improves your existing team during the off-season counts a player development.  While this may seem like the most common way for teams to improve, there is still a pretty wide variance on how well teams develop existing talent.

Free Agency: Bringing in existing elite-club talent from other teams to play for your team.  This could be a result of a job-change/move or just two buddies talking about playing together and then living the dream.

I guess the point of defining these three methods is so teams (particularly emerging or mid-level teams) can think about these methods and approach all of them during the off-season to make the most progress before practice begins.  There is also the side benefit of being able to look at a team's off-season in terms of these three modes of improvement, and really understand how some teams got better.  For example:

The biggest headline for the off-season was Bravo's acquisition of what felt like every free agent in the galaxy.  They got a big free agency signing last year bringing in Eric Johnson and Nick Lance, pouring it on with Kurt, Matzuka, Lokke, Keegan and your Mom just felt unfair.  But here is the thing, it just completed the trifecta of their off-season performance.  Bravo is traditionally a strong development team, getting quality points out of players that started young on that team and grew into star positions.  Then, they get a bomb draft class for national champion UC-Mamabird.

Everyone seemed ready to proclaim Bravo as the frontrunner for a national title and 8th seed in the NBA Eastern Conference playoffs and that kind of made sense.  But wait a minute, Revolver wasn't exactly sitting on its heels either.

The Moon Men are traditionally an insanely strong development team.  After all that was kind of the whole point to the team (Nick Handler, please tell me if I am wrong about this).  And they got good through development.  Sure, they hit a great free-agency class a few years ago as well, and that propelled them to the top, but don't discount how strong their development program has been.  Players like Little Buddy (Tim Gilligan) and Jordan Jeffrey feel like development successes for Revolver's program.  Revolver also had a strong draft class this year, pulling in more former-Polar Bear's players Eli Kerns and Simon Higgins.  I am biased, but both of those players a top notch players.  I guess you would need to include Cahill in the draft class since he wasn't on a team in the division last year.  He's not bad at ultimate and I hear he's a nice guy.

The whole point is that if you were looking at all three of these methods of improvement you would get a better sense of what is really happening during the off-season and not just big reactions to free-agency moves.  Bravo had a great off-season, but it wasn't like Revolver was staring at the stars the whole time.

Back to the first point rationalizing these three methods as pillars for off-season grading.  If you look at your team through the lens of these three modes of improvement you might learn a little about your team and be able to make some improvements you wouldn't have otherwise.  Let's go around the division and see what trends exist:

Ironside: strong free agency, good development and occasionally an excellent draft class

Sockeye: perhaps the best development in the nation(!), decent in terms of free agency and draft class

Doublewide: used to be a development heavy team, is still in the hangover from a previous free agency bomb and gets solid drafting from the giant state of texas

GOAT:  Great international drafting, I can't really speak to much else

Chain: Good free agency, baseline drafting and development (wait, isn't that my job . . . shit)

Ring: Typically strong development, middling drafting and free agency

Machine: good development, leaning on free agency this year

You could keep doing this for a while and if you did it for your team you might learn of a deficiency or strength you didn't already know about.  I don't think there is really anything all that special in coming up with these methods, but it might help start some conversations that end up being worthwhile.  Are you having a down drafting year?  put your chips into development.  Free agency class not looking so hot?  Scour the college to improve your draft class.  Back to watching Kill Bill 2, we're at the coffin scene.

Monday, July 07, 2014

Revisiting Film Technique

Holy crap there is a lot of video of ultimate available now!  It is like the first time I walked into a Toys 'R Us and I realized that a whole store could just have toys.  I didn't know where to look because everything was awesome.

With all of the video out there it now become so much easier to get content to support whatever it is you want to do.  Team scouting, instructional videos, highlight reels, etc.  But the technique of how to breakdown your film from a technical standpoint is still a barrier.  Outside of the coaching skill (and it is a skill that you can get better at) required to see what you need to see in the film, the process of clipping, telestrating and reforming film is tough.  I wrote about this for Worlds last year and it garnered a whole 2 comments!  So either no one reads this blog anymore (which is likely) or there aren't a lot of people really spending time on film study techniques.  I know that with the vast amounts of film out there most if not all elite teams are using film to scout and probably to improve their own game.  But the extent of that use might be pretty basic (throw the tape in the VCR and press play).

But since I have been doing a lot of work editing film I've learned a few more things and I thought I would post them here so future film-breakdown-wannabes will be able to stand on my shoulders and still not break 5'.


  • MPEG Streamclip has become my favorite editing tool again.  Mostly because of its superfast clipping mechanism.  Basically you watch film, press "I" and "O" to place clipping markers.  Then CMD+T (on a mac) trims the clip for you.  Save it in any format you want then the special thing is CMD+Z undoes the trimming and you have the full movie back.  This basically allows you to live-clip video the first time watching it, especially given MPEG's solid scrubbing tools.  There are some glitches, but it sure does beat my original technique (writing down time markers while I watched).
  • AVCHD is an awesome format for image quality, but it sucks for editing.  I've struggled with different AVCHD converters for the MAC.  They all feel like 1990s shareware and don't reliably get me the quality I want.  But it turns out that iMovie '11 can read AVCHD as a Camera Archive.  Then when you import the movie to iMovie it converts all of the clips in the AVCHD file to separate .MOV files.  The .MOV files get saved under your Movie Events folder so you don't have to actually "make" a movie in iMovie to get those files out.  Once you have the .MOV files you can go to town in whatever fancy editor you like
  • Explain Everything is an educational iPad app that has a really good recording feature.  I'm still green with getting it to work well, and its drawing tools aren't as great as other programs, but the ability to draw directly on the picture might push it into the front sport as the telestrator of choice.

There is plenty of space for growth in this field.  Ultimate is growing in so many other ways (membership, exposure, coaching), this is one way that isn't going to get attention for a while.  But eventually how well and quickly you can breakdown film might actually be a thing that kind of matters.

Monday, March 17, 2014

Mike Caldwell's Cutting Tree and "Cutting" as a skill

Recently Mike Caldwell wrote a response in Skyd about his cutting tree  Having recently been thinking about how to get our boys to be better technical cutters I was super excited.  Mike has been a premier cutter for a long time and if he was going to break down something like the route tree, that was going to be huge.  Often cutting technique in ultimate gets described as running harder for longer, and that isn't going to work against better athletes (of which there are plenty as you get older, trust me).  Mike is a great athlete, but he has also been a premier cutter for a long time and is a tireless workhorse.  Surely he has some insight on different ways to get open.

Unfortunately his piece, while having many excellent points (including one on sente which does a good job of explaining who should cut when), doesn't really describe how to cut better in general but rather how to run a certain pattern well.  The pattern to run is along the diagonals and sides of a trapezoid.  If you go back and watch Sockeye in the mid-2000s it is no surprise that this is the pattern Mike gives you.  Their use of isolated thermals as a cutting style to make an aggressively under-cutting alternative to Furious' H-stack was revolutionary and created a pattern that many college teams hope (and fail) to emulate with their offenses.  Mike's coaching points of what to think about, how to get better at the pattern are good insight and practice for young cutters.

I think the place that I take issue, or was disappointed by the piece was in the use of the term "tree."  The route tree in football is a description of the different paths that a receiver can take within a football offense.  The premise is that it contains all of the options for the receiver and that since both the quarterback and the receiver know the same tree they can be on the same page more easily.  With that in mind Mike's "tree" kind of holds all of the cuts that Sockeye ran (at least from their cutters) in the early-mid 2000s.  But it is more of a flow pattern and less a "tree."  In part that is a dilemma of continuous sports like ultimate being put in contrast to segmented sports like football.  There isn't a clearing pattern in football, there is a stoppage of play.  Maybe flow patterns are ultimate's equivalent to cutting trees, describing broad paths for players to take?  In that case, Mike's piece is a description of the old Sockeye flow pattern and is invaluable to players trying to be excellent cutters in that system.

However I think that cutting and flow patterns are different things.  Flow patterns tell you where the next cut should be and where you should go when you are done cutting (or aren't cutting at all).  Sockeye's H stack called for hard under cuts through the middle (often at an angle) and clears down the sideline (that could easily be deep cuts).  Cutting feels different.  Cutting is a move that is designed to create separation between you and your defender and is somewhat independent of the placement of the disc and more about the placement of the space you are trying to access.  I digress lest I spend the next 2000 words talking about space.


One of the big advantages of the football route tree is it tells you how to get where you are going.  It involves a discrete cut/movement to get open.  Five yards hard out then a 120 degree turn towards the quarter back (hitch).  Ten yards you then a 90 degree turn across the middle of the field (dig).  That is the element that I (perhaps naively) was looking for in Mike's piece and found missing.  I think I was hoping for a description of methods that Mike used to get open in different situations (with a defender fronting him, with a defender playing even, from a lateral reset, etc.).  What Mike provided really has only one (maybe two) cutting movements: a sharp change in direction at the top and another at the bottom of the trapezoid.  But that leaves a lot of different ways to get from A to B unexplained.

That is one of the other values of the football route tree.  Even if a particular offense doesn't use all of the tree (the Packers love their slants but don't throw too many flats), the branches are there for everyone to learn and understand.  It establishes the language of cutting in football.  Every community has a vernacular for their cuts, but the football route tree is almost universally consistent (sure there are subtle changes between things like a go route and a fade, but there is broad agreement on post, hitch, dig, out, slant, etc.).  We currently lack a common language for how to talk about these cuts.  What is a scoo or a whoop cut?  Is it a double cut or a triple cut?

So I think what I am going to spend some time on is developing a cutting tree that is about creating separation between you and your defender.  I'll talk to various coaches about different ways that they cut in hopes of generating the fundamental movement patterns/breaks that make up all of the cuts.  While many offenses don't use most of those cuts, coming up with a common vernacular and skill set to teach young players will hopefully help coaches develop talent faster and help players figure out multiple ways to get open to the same space.  This will be quite a little project, but I think there is some really work to be done there.  With that being said, what is your favorite cut?

Extra Note:  I've spoken with Kyle about this and I think he and I are going to work to make sure we are using the same names for cutting to start the ball rolling towards a common vernacular.  


Monday, March 03, 2014

TUFF vs UCF Coin Toss: What To Do With The Wind

I was reading an article by Jimmy Leppert on Skyd a second ago and was struck by some of the responses in the comments about a decision Texas had to make after the toss.  Apparently, and I hope I am getting this correct since it is all second hand, UCF won the coin toss and decided to take the up-wind side on a very windy day.  I guess kudos to Andrew Roca for choosing the correct side, but that was an easy choice.  Then Texas, coached by Calvin Lin who has been through his share of games, decides to start on offense.  Jimmy, the post's author, states that he was surprised by this decision and then people in the comments state that they would do the same thing.

So let's go over the rationale for receiving the pull going upwind to start the game:
-You can put your best players on the field going upwind with fresh legs
-You have to score upwind at some point anyway
-You hope that their defense isn't playing well early in the game
-You have the most amount of time to recover from the break that you give up by not scoring

I think most, if not all, of those fail to hold up to much scrutiny.  Let's just give up on the last two, because I think those are the easiest to dismiss (although if I need to in more detail I'll be happy to do so in the comments).

Let's focus on the 2nd on because that is the one that was being supported in the comments.  It is the one that seems to have the best chance of winning against scrutiny.  Texas DOES have to score an upwind point to win the game since the lost the coin toss, but not all upwind points are the same.  By receiving the pull Texas is likely to start at the brick (as the best scenario) or in the back of the endzone on the sideline (as the worst scenario).  Somewhere in the endzone is most likely.  Now Texas has to work upwind a full 70+ yards to score against a defense that is set and focusing on D.  That is a tall order even though you might have the best pieces in place (point 1) to accomplish that task.  UCF gets to set whatever defense they want and knows their assignments prior to the pull.

If that is how you are going to score your upwind point, the odds are pretty low.  Instead lets looks at an alternative: you start on defense.  This is the choice that Texas had and didn't take.  Here are some of disadvantages:
-"Weaker" defensive players on the field
-You have to get a break from the opponent
-They can always punt to force you to go 70 yards
-Pulling upwind you are likely giving the offense a shorter field to score

These seem like good reasons to go with the offense, but are they really?  The advantages of pulling are pretty decent too.  You get to control the tone of the point by setting the defense and as a result you can hope for an unforced error or block such that you get a short field.  That seems like the most compelling argument for pulling first.

If you pull you can try to get them stuck on a sideline and force either lateral throws to get off of the sideline or a straight punt.  At worst, if they punt, you have to work no more than 70 yards against a transition defense that may not have their assignments figured out yet (or might have a particular match up you can exploit).  At best you can block the punt or force backward passes to reduce your field even more.

So if you start with the pull you can at worst be in a situation better than if you started on offense, and at best have a situation that is better than all likely ones when you start on offense (I suppose there is a chance that your opponent will shank the pull and give you a short field, but those are super low odds).

If we go back through the list of advantages of receiving and think about them again I think the decision becomes clear:
-You can put your best players on the field with fresh legs: You can do that while starting on defense, and you can find a way to manufacture that later in the game if need be.
-You have to score upwind anyway: True, but not all upwind scoring opportunities are the same and the possible outcomes for an upwind score are better when you are starting on defense.
-You hope that their defense isn't playing well early in the game:  O.k. but if that isn't the case then their transition defense isn't likely to be playing well either.  Giving a slow starting defensive line the chance to play the defense of their choice with the opponent going upwind is a pretty friendly start to a game.
-You have the most amount of time to recover from the break that you give up by not scoring: True, but it would also be nice to not need that time.

I have ignored the mental aspects of starting down a break (likely 0-2 by the time the game gets back on track) mostly because those are so specific to a team's make up.  In general it would seem like you would never want your team to be in a position of weakness, but I have played against plenty of teams that seem to find strength in those positions.  The subject gets murky very quickly, so best to just leave that alone.

I guess my outcome is that it is almost always better to start on defense if you are forced to choose going upwind.  Maybe I should ask Kyle, I'm sure he's solved this already.